Studies in Jogless Stripes

[The Dreaded Jog]
[Meg's jogless jog]
[Meg's with a step back]
[Yellow stripe join]
[Four marked stripe joins]
I've been doing a little study on the best way to handle color changes, in preparation for a presentation I'll be doing soon for the Socknitters list members. In trying different ways to solve the problem of the dreaded color jog in circular knitting, I've hit upon what I think is a new way to handle it.

The problem of the jog

The problem, briefly, is that in circular knitting, you're really knitting in a continuous spiral, and the last stitch of any circuit around the tube comes out to be one row higher up than the first stitch.

It doesn't show in one-color knitted fabric, but in color patterning (the plain horizontal stripe being the simplest case) this shows up in the form of The Jog, at the point where the round begins. Please note that the jog actually happens on every round, in fact at every point on every round, but only shows where you change colors and can see how the rounds line up.

The old -- and still good -- way

The standard accepted best solution is often called Meg's Jogless Jog, though her mom talked about it too, and maybe it was known earlier (not in any book I have on my shelf, though). But Meg Swansen took it the farthest in her brilliant article in Knitter's #45 (Winter 96), where she showed how to apply it even in complex Norwegian two-color patterns. Meg's solution is when you have knitted the first round in your new color (or pattern) and arrive back at the first stitch of the new color, lift the stitch of old color from below the first stitch and knit it together with the first stitch. This in effect makes this stitch the last stitch of the first round, and shifts the beginning of the round one stitch to the left.

This is a very nearly perfect solution--in Meg's hands apparently absolutely perfect. In my hands, however, I get a sort of diagonal ripple or line of some sort, slanting up and to the left following the start of the round as it travels up the tube. It seemed to me that this might be caused by the fact that I was usually carrying the unused yarn up to the next stripe, as in a series of narrow stripes on one particular sock I was testing. See the photo of the sock leg.

The socks shown are a pair where I tested a different jog solution on each of the four "tubes"--the leg and foot of two socks. [The photo of the foot of the sock shows what happens if you use Meg's jogless jog, but don't shift the starting point leftward. Starting every round on the same stitch while knitting into the stitch below on every color change means you lose one round for every stripe, which can pull that column shorter, but it looks fairly neat.]

The New Idea

So, I wondered what would happen if I broke the yarn for every stripe, rather than try to carry the colors upward. The idea dawns: If I'm breaking the yarn anyhow, why should I attach the new color at the same place I ended the old color? Here:
If a stripe is truly jogless, then it's no longer a spiral and it doesn't matter where you start the next round!

And if you can move the beginning and end of each stripe around to any point on the tube, you can sandwich these troublesome points between two stripes that are perfectly smooth, with their own joins somewhere else. Use the ends of your yarns to fiddle your joins so that the rounds line up better, and there you have it, jogless stripes.

The downside: ends, lots of ends! Also, each stripe does structurally still have its own mini-jog, and it has to go somewhere. In this case you're wrenching the jog into line and it makes a little hump where the stitches are distorted. Except for a single-round stripe--where the tail meets the head on a perfectly level base and is totally invisible.

The lower two photos on this page show the new method. One's a detail of the join of the yellow stripe, the other shows four joins, to the outside of the tube from where the ring markers are placed. (And no, I didn't get those colors from a color wheel! It's a bad photo--digital cameras have trouble interpreting colors if there's no white in the picture--and leftover yarn anyway.)

What I'm working on now is discovering the precise way the ends have to be handled to best solve the jog problem.

Try it yourself!

Here's how you can do a test sample without knitting a whole tube--this is stockinette:

--Cast on about 20 sts and knit six or so rows in Color A, ending having finished a purl row.
--On the next row, slide half the stitches onto your right needle.
--Start knitting Color B in the middle of the row, purl back, turn and knit to the middle again.
--Cut off Color B and without turning the work, slide the stitches from the right needle onto the left.
--Pick up Color A and knit a few rows.

See what you can do to make the ends line up. Try different numbers of rows of Color B. See what happens with one row!

Okay, now you're where I am on this. You can start any row anywhere on your circle by sliding stitches around on your circular needle or by starting at a different one of your double-points.

Give it a try!


Go to String and Air
Go to my Home Page

Judy Gibson
Descanso, California
Email to jgibson (at) cts (dot) com
http://www.users.cts.com/crash/j/jgibson/string.htm